This repository contains the implementation of the ED-DenseNet model for gas-liquid two-phase flow pattern recognition.
data/: Contains sample images for different flow patterns.models/: Includes the model architecture definitions.scripts/: Training and prediction scripts.utils/: Utility functions and loss definitions.
Due to institutional confidentiality policies, the full dataset cannot be made publicly available. However, a representative subset of images is provided in the data/ directory. For access to the complete dataset, please contact the corresponding author.
- Training the model:
python scripts/train.py